Privacy Discussion Begins

I had the distinct pleasure Tuesday to sit in on a livestream of NIST Privacy Framework: Workshop #1.   Hosted by the National Institute of Science and Technology (NIST), Workshop #1 was the kick off of an initiative NIST is leading to develop a voluntary privacy framework.  Although the NIST Cyber Security Framework has been hugely popular across industries, NIST feels that it does not adequately address Privacy.  NIST’s objective is to establish “a voluntary Enterprise Risk Management tool that organizations can pick up and use to manage privacy risk.”  They have lofty goals that include producing a tool that can be used long into the future; encompasses emerging and unknown future technologies and uses of information; is as useful as NIST CSF; and to make the framework broad enough to be consistent with existing privacy and risk management standards, where practical! 

NIST recorded the three hour workshop and is going to make the recording available to anyone that wants to watch it.  I encourage you to do so as a lot of REALLY interesting concepts were discussed by some seriously qualified thought leaders in this space.  I’m super “geeked out” about this material and excited to share with you what I found most interesting.  None of this is final in any way but represents some of the conversations I found most compelling.

• Privacy is defined by the harm, if any, inflicted upon an individual by the way their information is handled.
• Harm is defined by each individual and may change over time.
• One individual’s harm may be different than another individual’s harm and is almost certainly different from the harm to the business that was the source of the privacy-related harm to the individual.

I personally think it is brilliant to be defining privacy in terms of the harm that it presents to an individual.  However, it has significant risk management ramifications that will need to be worked out in the privacy framework.

Risk Management Ramifications of the NIST Privacy Framework 

Identification of Privacy Risk

Organizations will need to know everywhere they have information about individuals.  The use of scanning tools will increase in order to find information across the enterprise.  But the information you are looking for may not be the obvious: name, address, account number, account balance, health information, etc., The question may be: what information do we have about an individual that could be used in a way that could bring about harm to an individual? You also have to ask, if we give any individual’s information to a third party, what could the third party intentionally or unintentionally do with the information that could harm an individual?  Will third party assessments begin to include questions to find out what other information third parties might have that can be combined with the information you are sharing with them, that could cause harm to an individual?

Inherent Risk Assessments

Defining privacy in terms of harm to an individual will make inherent risk assessments more challenging and scenario-based.  You will most certainly need to think outside the box to consider all the different ways information you collect and handle could harm an individual. How will you determine whether your information collection, information handling and sharing with third parties, potential breaches and incident response will harm any individual and by how much?  Will you need to start asking individuals how they would feel if their information was breached or used in an unintended manner?  Will your organization need to periodically refresh its understanding of individual harm, particularly as new technologies and uses of information emerge?

You will need to stay abreast of every new and changed way information is collected, managed, shared with a third party, destroyed, etc. In each of these cases you will no doubt need to document what and why information is being collected, the information lifecycle from collection to destruction, the intended use of the information, and the numerous possible uses of the information that could cause harm to an individual, including through your extended third party ecosystem. 

If you do conclude that information you handle could cause harm to individuals, how will you rate the risk?  What is the measure of harm – anything from financial loss, embarrassment, harassment, loss of time from unwanted marketing, black mail, psycho-social manipulation, even physical harm and death? Many of these kinds of harms do not readily translate in financial terms.

Residual Risk Assessments

With cyber security risk you apply appropriate organizational and technical measures to reduce the likelihood and / or impact of unauthorized access, alteration, or destruction of the information.  Defining privacy risk as harm to an individual(s), you aren’t solely concerned with unauthorized access, alteration, and destruction.  Your intended and unintended use of the information could cause harm. At a minimum, organizational controls will take on a relatively greater importance to ensure you are effectively capturing and controlling residual risk.

Risk Evaluation

Let’s say that you do find a way to rate residual risk in terms of harm to individual(s).  Mature organizations that manage risk against risk appetites and tolerances will have to go back and look at those values and somehow incorporate harm to individuals.  How much harm and what type(s) of harm to individuals will organizations be comfortable with?

Summary

NIST is just beginning the process to come up with a Privacy Framework and nothing is set in stone yet.  The privacy conversation is just beginning but it benefits each of us and our organizations to try and shape the conversation so any privacy framework published by NIST provides meaningful value without undue complexity and implementation heart burn.